relatively ethical: a comparison of linguistic research ...similarities between what at first glance...

Relatively Ethical: A Comparison of Linguistic Research Paradigms in Alaska and Indonesia

Gary Holton

University of Alaska Fairbanks

Just as there is no single model for community-based research, ethical standards for community engagement are not universal. Drawing from personal experiences with language documentation among threatened communities in two very different parts of the world, this paper examines the challenges of applying universal ethical guidelines for linguistic fieldwork.

1. IntRoduCtIon.1This paper contrasts my experiences with endangered language field-work in two very different regions of the world: Alaska and eastern Indonesia. As noted by Dobrin (2008), most analyses of endangered language fieldwork situations come from North America. So it is that most discussions of ethical practices in linguistic fieldwork rely almost exclusively on North American examples. For example, in a recent review of ethical issues in linguistic fieldwork, Rice (2006) deliberately restricts her attention to the North American context. To some extent this circumscribed approach reflects both academic ge-ography and political exigency. Whether explicitly acknowledged or not, North American fieldworkers are heirs to an Americanist linguistic tradition that continues to emphasize the role of Native speakers and community interaction. At the same time decolonization of the research paradigm in North America has led to a continual reexamination and reshap-ing of linguistic fieldwork models. In bringing to bear examples from my own fieldwork both within and outside North America, I hope to encourage an expanded discussion of fieldwork ethics that recognizes the diversity of linguistic fieldwork situations. The brief fieldwork vignettes presented in this paper clearly demonstrate that such diversity exists. Moreover, differences between fieldwork situations in Alaska and eastern Indonesia can have significant repercussions for our understanding of the application of ethical norms.

1 This paper was originally presented at a colloquium entitled Ethical Dilemmas Encountered While Documenting Languages: Examples and Responses, part of the 1st International Conference on Lan-guage Documentation and Conservation, March 12-14, 2009 in Honolulu, Hawai‘i. I wish to express my thanks to the colloquium organizer, Pamela Innes, and the participants in the colloquium. I also thank the three anonymous reviewers, who provided many useful comments and made me aware of important bibliographic references. Funding for the Tanacross Dictionary Project was provided by National Science Foundation grants 0136113 and 0332736. Funding for documentation of Western Pantar provided by National Science Foundation grant 0408448; National Endowment for the Hu-manities Documentation Endangered Languages Fellowship FN-50006-06; and a Hans Rausing Endangered Languages Project Field Trip Grant.

Vol. 3, No. 2 (December 2009), pp. 161-175http://nflrc.hawaii.edu/ldc/

http://hdl.handle.net/10125/4424

Licensed under Creative CommonsAttribution Non-Commercial No Derivatives License E-ISSN 1934-5275

Crucially, my focus here is on the interpretation and application of ethical principles rather than on the development or statement of ethical norms or guidelines. While frame-works for fieldwork ethics have evolved over the past several decades, a general consensus has emerged that emphasizes collaborative, community-based research. Witness the many introductory texts on linguistic fieldwork that include chapters devoted to ethics. Without exception, those texts published in the past decade echo broad principles of collabora-tive research based on working with speakers and communities (recent examples include Dwyer 2006; Crowley 2007; Bowern 2008). But while there is general agreement as to the overarching ethical principles of linguistic fieldwork, there is also a recognition that these agreed principles are subject to much interpretation. As Wilkins remarks, “the social, cultural, political, physical, and historical contexts within which linguists do fieldwork are probably more remarkable for their differences than their similarities” (Wilkins 1992:189). Rice is careful to qualify her conclusion with the caveat: “[T]here is not a ‘one-size-fits-all’ model, as different situations call for different types of interactions” (2006:150). The examples presented in this paper are an attempt to concretize such statements with actual examples from linguistic fieldwork in Alaska and eastern Indonesia. The two fieldwork projects that I will describe have broadly similar goals. Both fall under the rubric of endan-gered language documentation, and both attempt to produce similar products (e.g., diction-aries, audio-video recordings, pedagogical materials). Yet, achieving these similar goals in these two rather different language situations requires very different interpretations of ethical guidelines. In order to understand just why this should be, it is useful to begin with a brief overview of the language situations in Alaska and eastern Indonesia.

Alaska is home to some twenty or so indigenous languages, most belonging to either the Eskimo-Aleut or the Athabascan-Eyak-Tlingit families—both families of enormous geographic extent spanning much of the North American continent. Although the colonial history of Alaska differs significantly from that elsewhere in North America, the language situation in Alaska today is not so different from that elsewhere on the continent, due in large part to the political and social forces shared as one of fifty of the United States. In particular, all Alaskan languages are today highly endangered, with few speakers under the age of fifty, and most speakers of Alaska Native languages use English as their primary medium of communication.2

Eastern Indonesia is home to hundreds of indigenous languages. For the purposes of comparison, I will focus here on a single Indonesian province, Nusa Tenggara Timur (NTT). The province of NTT is home to some 70 or so languages, most belonging to the Austronesian family but also including a significant group of Papuan outliers (Grimes et al. 1997). The degree of endangerment of NTT languages varies greatly. Some, such as Nedebang (ISO 693-3 nec) on Pantar Island have few speakers below the age of 50. Others—including some that border Nedebang—are spoken by persons of all ages. There is also significant variation within languages depending on geographic location. Language shift, either to the national language, Indonesian, or to another indigenous language, has proceeded most rapidly among groups who have migrated to urban centers. Languages

2 In Alaska indigenous languages are more often referred to as Native languages. In this paper I use the terms “Native” and “indigenous” interchangeably.

A Comparison of Linguistic Research Paradigms in Alaska and Indonesia 162

LaNguagE DocumENtatIoN & coNSErvatIoN voL. 3, No. 2 DEcEmbEr 2009

remain strongest where speakers have remained in place on traditional lands. In general, in almost all areas of NTT outside of the urban centers, indigenous languages remain the languages of daily communication.

It should be emphasized that in spite of the obviously huge geographic and socio-cultural distance separating Alaska and eastern Indonesia, it is possible to point to several similarities. For example, both regions are home to great linguistic diversity, and both are threatened with rapid language shift due to the pressures of dominant non-indigenous pow-ers and the deleterious forces of development and mass communication. Eastern Indonesia is among the most linguistically diverse areas on the planet, and NTT is a meeting ground between the Austronesian language family and the Trans New Guinea family. Though ar-guably less diverse than Eastern Indonesia, Alaska forms a linguistic crossroads between the Asian and North American continents, serving as the intersection between the Na-Dene and Eskimo-Aleut families. Eastern Indonesia and Alaska also share in common a recent history of language shift, which may be attributed to forces largely outside the control of indigenous societies. While language shift (to English) is more advanced in Alaska, language shift (to Malay/Indonesian) in eastern Indonesia is following the same predict-able path followed by minority language groups that lack national political or economic power. In both NTT and Alaska, indigenous population represents less than one half of one percent of the national population. Educational and language policies are determined at the national level and explicitly promote national languages (and in Indonesia languages of dominant ethnic groups, such as Javanese) at the expense of minority indigenous lan-guages. Educational policy in NTT actively discourages use of indigenous languages. In Alaska the use of indigenous languages in schools was legalized only in 1972, by which time language shift to English was nearly complete. In both NTT and Alaska, television, radio, internet, and newspapers remain the exclusive domain of dominant languages. Such similarities between what at first glance might seem to be very distinct cultural regions reflect an underlying unity of the indigenous experience (cf. Smith 1999). However, there are also important differences between the language situations in eastern Indonesia and Alaska, and it is these differences that underlie the differing ethical norms to be discussed in this paper.

First, Alaska and eastern Indonesia differ remarkably in term of language attitudes. While Alaska does offer token support to indigenous language revitalization efforts, nei-ther Alaska nor Indonesia provides meaningful support for indigenous languages through official policy. However, attitudes towards bilingualism differ greatly in the two places. Indonesia is a multi-lingual nation; nearly every Indonesian of school age speaks more than one language. In NTT multilingualism is the norm, and most residents have command of a regional language of wider communication, a local Malay variety, and at least one neigh-boring language, in addition to their mother tongue. Many residents also gain fluency in other languages of NTT while working or attending school away from their home village. In contrast, Alaska has a long history of intolerance of non-English languages, dating back to transfer of colonial suzerainty from multilingual Russia to monolingual United States in the late nineteenth century (see Alton 1998). Indigenous Alaskans are constantly assaulted by the hegemony of the English language, not just indirectly via exposure to mass com-munication but also directly as a result of official policies such as English-only initiatives



that limit ability to use indigenous languages. This linguistic hegemony shapes Alaskan responses to ethical standards.

Perhaps the most notable differences concern the percentage of indigenous popula-tion as compared to the whole, and the presence or absence of an indigenous conscious-ness. The population profile of Alaska differs significantly from that of the other United States in that indigenous peoples comprise approximately 16% of the 670,000 residents of Alaska. No other state even comes close to this figure. Due in part to the uniqueness of this population, as well as to a shared history of colonial oppression, there is a strong sense of indigeneity among Native Alaskans.3 In eastern Indonesia indigenous persons comprise a much greater percentage of the population. While I lack precise statistics, it is clear that outside the provincial capital of Kupang and other urban centers, persons indigenous to NTT comprise the vast majority of the population. Perhaps because of this, NTT lacks a distinct indigenous consciousness. Indigeneity is the norm, not a separate identity. Displays of indigenous awareness are limited to celebrations of cultural traditions such as music, dance, and crafts.

Note also that the population of eastern Indonesia is vastly more rural than that of Alaska. NTT has a total population of over 4 million, but its largest city has a population of just 250,000. In contrast, approximately 60% of Alaskans live in urban areas.4 Moreover, almost all speakers of Alaska Native languages have easy access to the wider non-indige-nous community, including access to modern telecommunication services and transporta-tion services. For example, most can reach the urban center of Anchorage in a journey of a few hours or less. Fully half of the indigenous population—and perhaps a greater percent-age of fluent speakers of indigenous languages—resides in Anchorage and its environs.

Another important contrast between Alaska and eastern Indonesia concerns economy. Alaska is among the wealthiest US states (ranked 4 out of 50), and poverty rates for Alaska Natives are significantly lower than for Native Americans in the US as a whole (18.1 vs. 22.3%) (U.S. Census Bureau 2000). In contrast to Alaska, NTT is among the poorest of the Indonesian provinces. Most of the population outside urban centers rely almost entirely on subsistence farming and fishing, with no cash income.5 All of these factors conspire to yield the ironic observation that, in spite of the fact that Alaska has vastly lower population density than eastern Indonesia (0.4/km2 vs. 89/km2), Alaska’s population is actually much more urban. Alaska’s indigenous population is much more integrated into the larger non-indigenous population, with easy access to transportation, communication, and the cash-

3 This indigenous consciousness is reflected in the existence of numerous organizations representing indigenous people, such as the First Alaskans Institute, the Federation of Alaska Natives, and the Inuit Circumpolar Council-Alaska.

4 This figure was arrived at by using census data for the three urban boroughs: Municipality of An-chorage, Fairbanks Northstar Borough; and the City and Borough of Juneau.

5 While subsistence continues to play a significant economic role in some rural Alaskan households, its role is largely overshadowed by the cash economy. Nevertheless, subsistence hunting and fishing has become an important identity marker for indigenous Alaskans.



based economy. As a result, indigenous Alaskans are able to participate in the larger North American dialogue regarding rights, property, language, and economic gain.

In summary, the majority of indigenous peoples in eastern Indonesia reside in rural areas, living a subsistence lifestyle, and continue to speak ancestral languages. The major-ity of indigenous peoples in Alaska have undergone urbanization, participate in the cash economy, have greater access to mainstream North American society, and have (been) shifted away from indigenous languages.6 While this brief summary grossly oversimplifies the complex social and economic fabric of the two regions, these characteristic differences between Alaska and eastern Indonesia help to contextualize the difficulties encountered in attempting to apply principles of ethical fieldwork in those two regions.

2. APPLyIng EthICAL PRInCIPLEs. When we speak of ethics in language documenta-tion we generally mean normative ethics—that is, how one ought to behave in fieldwork situations. The problem of course is that there is no single norm—no single universal stan-dard for what is morally “right” or even a single set of best practices in the conduct of language research. In order to have broad application, putative universal guidelines are necessarily vague, and the application of such broad principles as “do no harm” or “work collaboratively” requires interpretation relative to a specific cultural context. In the absence of such context, what is considered ethical in one part of the world may fail to be seen as ethical elsewhere.

To illustrate this point I provide here examples of two types of ethical issues that I have faced in both Alaska and Indonesia. In an earlier version of this paper I referred to these issues as ethical dilemmas, but strictly speaking these are not ethical dilemmas, for they do not involve conflicts (real or apparent) between competing moral imperatives or ethi-cal guidelines. Rather, these are dilemmas of interpretation, in which an action that might be interpreted as ethical in one fieldwork context may be considered unethical in another context, and vice versa.

These dilemmas are not necessarily the most significant or important issues I have faced in my fieldwork; rather, they are intended as mere case studies that illustrate the relative nature of ethical practices. The Alaskan example is drawn from my work with Tanacross, an Athabascan language spoken by about thirty people in eastern interior Alaska (ISO 639-3 tcb). Tanacross was the subject of my doctoral research beginning in 1997. The Indonesian example is drawn from my fieldwork with Western Pantar (also known as Lamma, ISO 639-3 lev), a non-Austronesian language spoken on the island of Pantar in the eastern Indonesian province of Nusa Tenggara Timur. I have been working on the documentation of Western Pantar since 2004. Crucially, these examples involve fieldwork practices that are directly comparable across the two different regions.

2.1 on thE notIon of PRofIt And EConomIC gAIn. In at least some parts of the world—including North America—the issue of economic gain is a touchy one for docu-

6 Given a long history of deleterious Native language policies and current policies that promote Eng-lish-only, it might be more accurate to phrase this with less agency and say that indigenous peoples in Alaska have been shifted away from indigenous languages.



mentary linguists. In Alaska there is an overarching concern and wariness that outsiders (non-community members) may be profiting from language work—in particular from sales of materials derived from language work. Part of this feeling stems from resentment at the fact that linguists owe their academic careers and appointments to the Native community. In the words of an email I received recently: “Academics take and take and take but what do they give back? Possibly a dictionary that will gather dust except for the very few that will utilize them” (anonymous, pers. comm.). This message reflects an established ideol-ogy that equates research with taking from the community (Smith 1999). It is tempting to deflect such accusations with the justification that linguists working with Native languages could just as easily be employed working instead on English or some other non-Native, non-endangered language. Yet in truth it must be acknowledged that many linguists do indeed owe their careers to Native languages, whether directly or indirectly. But the kind of profit of which I speak here is a more tangible one. Not the salary I earn as an academic linguist but rather the potential profit I can supposedly make selling Native language books and recordings.7

The perception of potential (if unrealized) profits often lies behind desires to restrict outside access to language materials. This notion of profit goes beyond the wages earned as collaborators in a language project. Even when participants are directly compensated for their work (as indeed they have been in both my Alaskan and Indonesian fieldwork), there is often concern about the loss of potential communal profits from the language mate-rial itself. This was made very clear during my recent work with one particular language group in Alaska to develop a local language archive. The community wanted to ensure that none of the stories recorded as part of the project could be used by other authors for profit. In this case the concern was about English translations of stories recorded in Native language—a tacit acknowledgement that potential profits lie in the English version, not the Native language version. But the message was clear. The language archive represented a source of potential wealth: not just cultural wealth, but potentially real wealth in terms of economic profit. Within the Alaskan and wider North American context profit is viewed as a zero-sum game; that is, profit by an outsider is viewed as a loss by the community even when there is no direct loss of income to the community (though there may be real but in-tangible losses). This is because when an outsider makes a profit from language materials, the community may lose the potential for future income from those materials. It may be the case that the community itself does not possess sufficient resources to make a profit from the material, but even so, the community may be foregoing potential licensing fees in order to allow the outsider to make use of the materials. This foregone profit is objectified by the community as an economic good.

But attitudes toward profit by others vary widely among cultures. In part, this variation stems from the fact that the correlation between absolute wealth and quality of life differs across cultures. Put another way, “international comparisons of happiness and income are significantly influenced by cultural factors” (Easterlin 1995:42). In particular, sensitivity

7 Whether or not such profits are in fact possible is another issue. I assume here that they are not, though I sometimes wish there were a way to make a profit selling Tanacross language materials so that we could use those profits to support language maintenance efforts.



to income disparity may vary widely among different cultures (Diener and Oishi 2003). Within modern American society—including Native Alaska—there is a high degree of sensitivity to income disparity: in experiments most Americans indicate a preference for relative wealth over absolute wealth. That is, North Americans prefer to be relatively richer than their peers even if it means that they (and everyone else) are less rich.

In contrast, Diener and Oishi (2003) show that within less wealthy societies there is a stronger correlation between wealth and subjective well-being, which tends to over-ride concerns about relative wealth. While extreme cases of income disparity such as the excesses of the Suharto family may be criticized, on the whole income disparity is toler-ated in Indonesia so long as it leads to overall gains in absolute wealth.8 This is precisely what I have experienced in my fieldwork in eastern Indonesia. Rather than people being concerned that I might profit from language work, there is an expectation that I will profit. There is a general assumption that I am selling CD’s and books back in America, teaching people how to speak Western Pantar. Crucially, there is an equal expectation that profit will flow back to Pantar via patronage. Just as Dobrin (2008) notes for nearby Papua New Guinea, exchange also plays a crucial role in eastern Indonesian society (van Wouden 1968; Fox 1980). In traditional Alor-Pantar society, wealth was never used to accumulate material goods but instead used solely to cement exchange relationships (Du Bois 1944). While this may be less true today, exchange and redistribution remain an important obliga-tion of wealthy persons.9 As a foreigner from a relatively wealthy society—and thus by default a “big man”—I am expected to develop extensive exchange relationships with the language community through patronage. Under this view there should be an increase in absolute wealth for all parties: both the outside linguist (through putative sales of CD’s in America) and the community members (through patronage). Disparities in relative wealth are of lesser concern. As the “big man,” I am expected to profit more in relative terms, but as long as sufficient patronage is supplied, it need not be at the same absolute level.

In the eastern Indonesian context, failure to profit can lead to some awkward moments. About a year ago I went to Indonesia to complete a dictionary of Western Pantar. My co-author and I worked for a couple of months finalizing the contents, doing final edits, and formatting the book layout using a desktop publishing program. We then made arrange-ments to travel to Kupang, the provincial capital, to have the book printed. Along the way we stopped in the regency capital and sought printing subventions for the dictionary from the regent.10 He graciously supplied us with a nice bundle of cash to be used for printing,

8 Suharto was president of Indonesia from 1967 until his resignation in 1998 amid accusations of widespread corruption. It is estimated that Suharto embezzled more money than any other govern-ment leader in history (Transparency International 2004).

9 Unfortunately, ethnographic studies of modern Alor-Pantar society are severely lacking. Those that do exist focus on coastal regions that, as noted by Barnes, are “sharply distinguished in culture and language from the more indigenous mountain population” (1975:349).

10 A regency (Indonesian kabupaten) is an administrative unit smaller than a province but larger than a district (kecamatan). A regency is headed by a regent (bupati).



with the understanding that half of the printed copies would go to him to distribute as he saw fit. When it finally came time to pay the printer, my co-author was incredulous when he discovered that I intended to spend the entire amount given by the regent. That is, I would receive no profit. I told the printer how much I had to spend, and the printer told me how many copies he could print. When the books were delivered a few weeks later, I gave half to the regent and distributed the remaining half in the village. There was no “left-over” profit that could be redistributed through exchange. This was probably not the right thing to do, for it left me with no profit that I could use to distribute patronage to the community. Rather than keeping a portion of the regent’s money for myself (and via patronage, the community), I had spent it all on book printing.

I should emphasize that this was not a question of money per se but rather a question of profit. The level of my financial commitments in Pantar dwarfs any potential skimming from the publishing subventions. While it would have been inappropriate to pocket a large portion of the funds, keeping a small portion, say ten percent, would have been more than acceptable. In retrospect this now seems obvious to me, for this is the normal way in which funds are distributed in the region. For example, a school headmaster in Pantar receives monthly finances for his school in a lump sum; anything left over after payment of teacher wages and other expenses can then be retained as profit. In my case retaining ten percent of the publishing subventions would have made little real difference to the community even if the profit had been completely redistributed to the community. I had already developed a much deeper financial commitment to the community as an employer of research assistants, a renter, and a hirer of porters and motorbike drivers. And in any case the printed books were themselves distributed to the community at no cost. (Those books themselves could be viewed as patronage.) Without discounting the value of the money itself in real terms, I would argue that my failure to profit was awkward more because it clearly demonstrated my ignorance of (or disregard for) an established system of exchange and patronage. I was not acting as a “big man” should.

2.2 CommunIty-bAsEd REsEARCh And thE notIon of CommunIty. One theme that emerges repeatedly in discussions of ethics is the need for what I will refer to here as community-based research (CBR). This approach has been referred to by a variety of names, including: collaborative research, action research, and participatory research (cf. Harrison 2001; Cameron et al. 1992; Dwyer 2006). It is of course possible to make finer distinctions among these various types of research paradigms, and each of these labels may take on more specific meanings when applied to a particular discipline. However, all of these approaches share in common a focus on community engagement. They presume an ongoing dialogue between the community of speakers and the researcher—whether the researcher comes from within or from outside of the community.

The word “community” is raised often in discussions of language documentation. For example, it occurs in 17 of 96 titles of presentations given at the 1st International Confer-ence on Language Documentation and Conservation, and occurs fully 329 times in the



abstracts of those presentations.11 This emphasis on community reflects current fieldwork practice. As stated in a recent linguistic fieldwork manual, “there is an overriding ethical requirement that the general aims of the projects should be understood by the community … as well as accepted by the people in that community” (Crowley 2007:70). Clearly, the concept of community is felt to be important to language documenters. Unfortunately this important concept is treated as an undefined, abstract entity.

The obvious difficulty is that there are many levels of community, so that it is not easy to determine who speaks for or represents the community. The answer may depend on who asks the question. Viewed from afar (e.g., an NSF grant reviewer), a regional language center may be sufficiently representative. On the other hand, when viewed from within a community, a regional language center may be seen as distant and removed, staffed by people who were not raised in the community and have not been vetted by community leaders. As an example, consider the following many different possible views of the Pantar language community:

regional language center in the provincial capital of Kupang• group of Pantar speakers in the provincial capital of Kupang• group of Pantar speakers in the regency capital of Kalabahi• NGO in regency capital formed by Pantar speakers but working on • development issues“mayor,” “teacher,” or other official person in the regency capital• “mayor,” “teacher,” or other official person in the village• a particular village or villages• engaged and interested speaker•

Any of the above could legitimately be considered to represent the Pantar community, depending on the context or perspective of the project. Another recent fieldwork manual acknowledges this difficulty, cautioning fieldworkers to “make sure that you are seeking permission from the right people” (Bowern 2008:153). Carrying out such relativistic rec-ommendations requires a significant amount of finesse on the part of the researcher. Unfor-tunately, most ethics guidelines ignore context and instead adopt an easily operationalized solution that favors “official” representatives.

In both the North American and Indonesian contexts there is no shortage of candidates for official representatives. North American academics must have their research propos-als vetted by university ethics review boards, many of which now include representatives from the indigenous communities in which the research takes place.12 Foreign researchers in Indonesia must seek permission from the Indonesian Academy of Sciences. But neither university ethics boards nor national scientific organizations necessarily have the ability—let alone the desire—to look out for the needs of local communities. Recognizing this

11 Abstracts available at http://nflrc.hawaii.edu/ICLDC09/abstracts.html, accessed March 20, 2009. The latter number consists of the plural form as well as the singular.

12 Ethics boards are known in the United States by the term Institutional Review Board (IRB) and in Canada by the term Research Ethics Board (REB).



problem, many authors exhort language workers to seek the approval of local language committees, village councils, or local research councils (cf. Assembly of Alaska Native Educators 2001). However, in many cases the speakers with whom we work may not feel well-represented by the official leadership—or even the local leadership. Moreover, rely-ing on official leadership can invite caprice. In small Alaskan villages where family feuds have raged for decades, a change in the official leadership of a village council can lead to a revocation of a researcher’s welcome in a community. At its very worst, official deference risks complicity in a corrupt and abusive system.

As a way of comparing approaches to community in Alaska and Indonesia we can compare two specific language documentation projects. Both involve the production of a dictionary. Both involve community, though in very different ways. The Tanacross Dic-tionary Project began in 2003 as an outgrowth of a summer language revitalization work-shop known as the Athabascan Language Development Institute. Students at the Institute expressed frustration at the lack of a dictionary for Tanacross. Indeed, Tanacross was one of the only Alaska Native languages for which no dictionary had ever been compiled or printed. The best we had was a lexical card file that made use of a long-abandoned practi-cal orthography. So the participants in the language institute resolved to create a dictionary that would be useful for students of Tanacross. We were a motley crew consisting of elder Native speakers, a certified Native language teacher, younger language learners, a linguis-tics graduate student, and me. We were not officially selected or vetted. Indeed, we were self-selected by our interest in Tanacross language and our willingness to commit three weeks of the brief Alaskan summer to language learning. We sought funding from the US National Science Foundation and set out to design a dictionary through a series of broader community workshops. These workshops were sponsored by the Tanacross village council and were attended by both the core group from the language institute as well as many other community members. We held additional workshops in Fairbanks under the auspices of the Tanana Chiefs Conference, a regional Alaska Native corporation. In very many respects the Tanacross Dictionary Project was a grass-roots effort that took a bottom-up approach to community.

In contrast, the Pantar dictionary project was not really a project at all; it just kind of happened. In 2004 I set out to assess the possibility of documenting Western Pantar, a language about which neither I nor anyone else outside of Pantar knew very much about at the time. This was not at all a grass roots effort. I had never been to Pantar and had no reason to believe that the community had an interest in language documentation. Instead, the motivation was external, based on (1) my own idiosyncratic interest in Papuan outliers; (2) the fact that the extant documentation of the language consisted of a 117-item wordlist; and (3) intriguing reports that suggested the presence of typologically unusual argument structure. This was clearly a top-down approach to community.

The idea for the Pantar dictionary came from my collaborator Mr. Mahalalel Lamma Koly, a self-taught linguist who has been documenting his native language as a hobby since he was in high school. It was Mr. Lamma Koly who sparked my own interest in lexicography. As a documentary linguist I was of course dutifully recording a lexicon, but I must admit that I started with a personal (and perhaps misguided) bias toward grammar and discourse and argument structure alternations. Together Mahalalel and I dived into the intricacies of technical vocabulary: names for birds; verbs for different types of cook-



ing; types of rocks; species of millet. As we moved into more and more technical areas of vocabulary we were joined by more and more experts. Evenings at the Lamma Koly household became the locus of heated debates about household construction terminology and dryland rice landraces. There was no official language committee, no official team. But everyone in the community was soon aware that a dictionary of Western Pantar was in the making, and Mahalalel’s house was the place to be if you wanted in on the action.13 (It probably helped that Mahalalel’s house was also the only place in the village with electric light, thanks to a solar panel I had brought with me.) More often than not, these discus-sions were monolingual—something that could never have happened in Tanacross, due to the greater degree of language decay.

Mr. Lamma Koly continued to work on the dictionary after I returned to Alaska, con-tinuing to seek advice from others in the village. I returned to Pantar again in April 2008, and we began the process of editing, formatting, and printing the dictionary. We began by entering the data that he had collected in notebooks during the time I was away. Then we began to systematically edit those data, often expanding on particular semantic domains. Again, as with the initial data collection, the editing process was top-down, and most of the editing work was done by Mahalalel and me in Mahalalel’s house. But in truth no one ever works alone in Pantar society. Mahalalel’s house remained an important gathering place, and it was a rare occasion when Mahalalel and I found ourselves alone in the usually active front room of Mahalalel’s house. More often men and women (well, mostly men) were gathered around the room participating in discussions about word meanings and helping us to expand certain semantic domains. Some of these were casual visitors, whereas, others became regular participants. Some had only general comments to offer, while others were specialists in certain domains—for example, fish terminology. But all had something to contribute.

As the brief outlines above demonstrate, both the Tanacross and Pantar dictionary included components that could be considered to be community-based. Comparing the two projects we can ask which dictionary was more community-based. The answer is not immediately obvious. In spite of the grass-roots beginnings of the Tanacross dictionary, as time progressed the dictionary came to be seen as an individual project as opposed to a community effort. We had funding to pay only some elders for recording, with the result that only four voices appear on the recordings that accompany the dictionary. And most of these recordings are from a single speaker: Irene Arnold, the language teacher and princi-pal collaborator on the dictionary. Perhaps the funding model led to some of the original participants feeling excluded from the project. Tellingly, Irene Arnold is listed as first au-thor on the published dictionary (Arnold, Thoman, and Holton 2009), and the dictionary is often referred to by community members as “Irene’s dictionary.”

13 A reviewer commented that this tendency for people to congregate around whatever project hap-pens to be going on may reflect a wider cultural pattern associated with a rural agricultural society. That may well be the case. However, while it is true that there are relatively fewer entertainment options in Pantar as compared to Tanacross, it is not the case that Tanacross residents were too busy with other tasks to participate in the dictionary project.



In contrast, at its inception the Pantar dictionary didn’t really have any grass roots support. It was very much a project by Mr. Lamma Koly and me. And yet, in the process it was adopted by the community. It became a community-based project. No one received payment for supplying words or definitions for the dictionary. Rather, words were shared for the pure joy of discussing the intricacies of the language, marveling at the richness of Pantar biotic vocabulary in comparison to Indonesian. Men drew sketches to indicate names for parts of houses. Women demonstrated different cooking techniques. Children joined the effort with captured birds and other animals for identification. Others contrib-uted to the final production, for example, by supplying woven cloth to be imaged for the dictionary cover. The level of communal involvement in the dictionary project reflected a deep level of communal ownership of the language, one that transcended any one fam-ily or single village. Although my name is listed as first author (Holton and Lamma Koly 2008), there was never any doubt whose language this was. As far as the community was concerned, this was “our language.”

These differing experiences of community in the Tanacross and Pantar dictionary projects reflect differing linguistic ideologies and differing conceptions of the individual within society. What this means for our understanding of ethical fieldwork is that what counts as “community” must ultimately be defined on an ad-hoc basis, relying on personal relationships and a genuine commitment to the language and its speakers. Even where of-ficial notions of community can be more readily identified, these notions tend to be more relevant in the early stages of fieldwork, when personal relationships are still being worked out. This leads us back to the founding of linguistic fieldwork as an enduring relationship between linguists and native speakers. In the recent rhetoric, collaboration is often viewed as an explicit rejection of colonialist research that failed to acknowledge indigenous people as equal participants (cf. Harrison 2001). However, we may have this precisely backwards. Rather than being a reaction to colonialist practices, collaboration and community-based research may equally be seen as the natural evolution of traditional fieldwork practices. To the extent that linguists do actually choose to engage in community-based research, I believe they do so more often not because it is seen as more ethical but rather because it is seen to produce better results. Indeed, Rice (2006) identifies the genesis of the current collaborative framework for linguistic fieldwork in early attempts to improve the linguist-centered model, and Dwyer (2006) observes that the “lone-ranger” approach to language documentation is simply inefficient over the long term.

3. ConCLusIon. What these two case studies show us is that what counts as ethical re-search may vary across cultures and languages. In Alaska the notion that a linguist might profit from linguistic research is anathema. Impressions of profit are avoided at all costs by repeatedly stressing that the work is being conducted in collaboration with the community, and by giving copies of publications to indigenous collaborators to distribute for their own profit. In contrast, to avoid or forgo profit in Pantar was viewed by my co-author as at least culturally inappropriate, if not downright unethical.

Similarly, the requirements of community-based research differ greatly between Alas-ka and Pantar. It would be inconceivable to have begun the Tanacross dictionary project by working with a single speaker or small group of speakers. Rather, the Tanacross dictionary project emerged after a long process of discussion among speakers, learners, and linguists.



As the years wore on the core of participants dwindled in number, but this was acceptable, as the project had already been vetted by the community as a collaborative process. The project had already met the ethical requirement of community engagement. In contrast, in Pantar there was no ethical exigency to begin the dictionary project with a sequence of community discussions. Indeed, the project might have been viewed as less legitimate had we abdicated our role as “experts” and instead sought community advice.

The possibility remains that I have simply misunderstood the ethical norms and expec-tations in Alaska, in Indonesia, or in both places. This seems especially likely in the case of Indonesia, where I am a non-native speaker of both the indigenous language and the con-tact language, and the dialogue of indigenous research is less fully developed than in North America. Nevertheless, the two examples discussed in the paper—two research anecdotes, if you will—suggest that the pursuit of universal interpretations of ethical guidelines and standards may prove elusive, or in the worst case even harmful. In other words, what is good for Alaska may not be good for Indonesia, and vice versa.



RefeRences

aLtoN, thomaS. 1998. Federal policy and Alaska native languages since 1867. PhD dis-sertation, University of Alaska Fairbanks, Fairbanks.

arNoLD, IrENE, rIcharD thomaN, and gary hoLtoN. 2009. Tanacross learners’ dictionary: Dihthâad xt’een iin anděg dínahtlǎa’. Fairbanks: Alaska Native Language Center.

aSSEmbLy of aLaSka NatIvE EDucatorS. 2001. Guidelines for strengthening indigenous languages. Fairbanks: Alaska Native Knowledge Network.

barNES, robErt h. 1975. Comments on a Pantar language. Bijdragen tot de Taal-, Land- en Volkenkunde 131(2/3):348–350.

bowErN, cLaIrE. 2008. Linguistic fieldwork: A practical guide. New York: Palgrave Mac-millan.

camEroN, DEborah, ELIzabEth frazEr, PENELoPE harvEy, m.b.h. ramPtoN, and kay rIcharDSoN. 1992. Researching language: Issues of power and method. New York: Routlege.

crowLEy, tErry. 2007. Field linguistics: A beginner’s guide. Oxford: Oxford University Press.

DIENEr, ED, and ShIgEhIro oIShI. 2003. Money and happiness: Income and subjective well-being across cultures. In Culture and subjective well-being, ed. by Ed Diener and Eunk-ook M. Suh, 185–218. Cambridge: MIT Press.

DobrIN, LISE m. 2008. From linguistic elicitation to eliciting the linguist: Lessons in com-munity empowerment from Melanesia. Language 84(2):300–324.

Du boIS, cora. 1944. The people of Alor: A social-psychological study of an East Indian island. Minneapolis: University of Minnesota Press.

DwyEr, arIENNE m. 2006. Ethics and practicalities of cooperative fieldwork and analysis. In Essentials of language documentation, ed. by Jost Gippert, Nikolaus P. Himmel-mann, and Ulrike Mosel, 31–66. The Hague: Mouton de Gruyter.

EaStErLIN, rIcharD a. 1995. Will raising the incomes of all increase the happiness of all? Journal of Economic Behavior and Organization 27(1):35–47.

fox, JamES J. (ed.) 1980. The flow of life: Essays on eastern Indonesian. Cambridge: Cam-bridge University Press.

grImES, charLES E., tom thErIk, barbara DIx grImES, and max Jacob. 1997. A guide to the people and languages of Nusa Tenggara. Paradigma, series B 1. Kupang, Indonesia: Universitas Kristen Artha Wacana and Alfa Omega Foundation.

harrISoN, barbara. 2001. Collaborative programs in indigenous communities: From fieldwork to practice. Walnut Creek, California: Alta Mira Press.

hoLtoN, gary, and mahaLaLEL Lamma koLy. 2008. Kamus pengantar bahasa Pantar Barat. Kupang, Indonesia: UBB-GMIT.

rIcE, kErEN. 2006. Ethical issues in linguistic fieldwork: An overview. Journal of Aca-demic Ethics 4:123–155.

SmIth, LINDa tuhIwaI. 1999. Decolonizing methodologies: Research and indigenous peo-ples. London: Zed Books.

traNSParENcy INtErNatIoNaL. 2004. Global corruption report 2004. London: Pluto Press.u.S. cENSuS burEau. 2000. American FactFinder Census 2000 Demographic Profile High-

lights. Online: http://factfinder.census.gov (Accessed 11 March 2009).



wILkINS, DavID P. 1992. Linguistic research under Aboriginal control: A personal account of fieldwork in Central Australia. Australian Journal of Linguistics 12:171–200.

vaN wouDEN, f.a.E. 1968 [1935]. Types of social structure in eastern Indonesia, translated by Rodney Needham. The Hague: Marinus Nijhoff.

Gary Holton

[email protected]



relatively ethical: a comparison of linguistic research ...similarities between what at first glance...

Documents